LONDON — Previous research has shown that providing a clinically definite multiple sclerosis (MS) diagnosis often takes several weeks or even months. But new research suggests that a new "technology platform" can help provide an accurate diagnosis much sooner by identifying gene expression patterns in blood samples.

The IQIsolate (IQuity Inc) system consists of a "suite of algorithms" that analyze blood-based RNA markers.

The pilot study showed that a machine-learning algorithm created by IQIsolate, and based on more than 700 peripheral whole blood samples, distinguished patients with MS from healthy volunteers and from patients with other neurologic disorders with more than 90% accuracy.

In addition, the algorithm "captured" patients with clinically isolated syndrome (CIS) who later progressed to clinically diagnosed MS.

"Basically, we've identified genes in patients with MS that are different from other inflammatory and noninflammatory diseases," lead author Chase F. Spurlock, PhD, Vanderbilt University School of Medicine, Nashville, Tennessee, told Medscape Medical News.

"And we've been able to develop an assay that's reproducible and greatly accurate in identifying patients with MS at the earliest clinical timepoints." Dr Spurlock, who helped develop the testing system and is a shareholder in IQuity, presented the study results here at the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS) 2016.

The test, which is going through the Clinical Laboratory Improvement Amendment Certificate of Waiver process, is expected to be launched in the United States within the next year.

Machine-Learning Methods

The investigators note that the gold standard for diagnosing MS is currently MRI "detecting damage to the central nervous system (CNS) in the form of spinal cord and brain lesions disseminated in space and time."

But "tools capable of detecting inflammatory processes leading to CNS damage would be of considerable clinical utility since early detection and initiation of therapies lead to optimal patient outcomes," they add.

IQIsolate uses machine learning methods from worldwide patient samples to distinguish between autoimmune diseases through differences in gene expression.

Dr Spurlock noted that Vanderbilt has been examining gene expression differences for more than 10 years in hopes of developing clinically useful diagnostic tools. "What's interesting about this approach is that it only requires a tube of blood. And it's a minimally invasive process," he said.

The study included blood samples collected from 376 healthy participants (including 178 with a family history of MS), 267 patients with MS, and 76 patients with CIS.

A sample of 125 patients with other inflammatory and 116 with noninflammatory disorders was also assessed.

"Differentially expressed protein-coding genes and noncoding genes" were identified from the sequencing data by using TaqMan (ThermoFisher Scientific) gene expression arrays.

Testing and retesting led to the creation of the disease-identifying algorithm. In addition to showing multicategory classification of the patients with MS or other neurologic disorders and of the healthy control group, "we found that a portion of the MS patients…exhibited elevated levels of interferon response genes," report the investigators.

"Promising" Findings

"What was impressed upon me here at ECTRIMS was the need for other tools that doctors can turn to," stated Dr Spurlock. "We're not trying to go after the MRI, we're just offering another diagnostic tool for physicians."

He said that the goal of the investigation was to see whether a patient with MS looks like one with a different neurologic disease. "And they do look similar. But we could still pull them out with a greater than 90% accuracy."

Dr Spurlock reported that when the test is launched in the United States, it will be available only to clinicians and not direct to consumers. When asked how easy it will be for clinicians to interpret the test's findings, he answered that results will show "a simple yes or no for a gene expression signature consistent with multiple sclerosis."

"We're not diagnosing MS. That's their job," he added.

Jacob McCauley, PhD, Department of Genetics at the University of Miami, Florida, commented to Medscape Medical News that the investigators "did a lot of good work and provided a lot of interesting calculations" in their poster.

Dr McCauley, who was not involved with this research, added that the findings hold promise.

"There's a lot of work that still needs to be done, but this could be promising. You need to do a lot more sampling and modelling and training sets, but there's some hope there," he said.

The study was funded by the National Institutes of Health and by IQuity Labs Inc. Dr Spurlock and two of the four other study authors are shareholders of IQuity Labs.

Congress of the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS) 2016. Late-Breaking News Poster P1652. Presented September 16, 2016.

Follow Deborah Brauser on Twitter: @MedscapeDeb. For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.